Mice and rats are a fundamental component of drug discovery research. Standard outbred stocks and inbred strains of rodents are used extensively for drug safety and pharmacokinetic testing. Our ability to generate specific mutants through transgene insertion or gene knockouts has greatly expanded their utility.
The NIH has begun the Knock-Out Mouse Project (KOMP), an ambitious program to knock out every gene in the mouse genome. These knockouts are extraordinarily useful for modeling human diseases and elucidating key metabolic and regulatory pathways.
These valuable animals have been widely disseminated throughout the research community. Unfortunately, there is a widespread perception that the phenotypes displayed by these animals are attributable to the genetic manipulations they carry. This is not necessarily true. No gene acts alone, but in concert with all the other expressed genes in the organism. Even classic single-gene Mendelian traits, such as sickle cell anemia, display phenotypic variation due to the modifying effects of other genes. Thus the genetic background must be considered in these studies. Rapid, cost-effective, and comprehensive genotyping is essential.
Mice and rats fall into two broad classes of genetic background. Inbred strains of animals have been generated by repeated sibling mating so that all the animals in an inbred strain are considered genetically identical. Outbred stocks of animals are managed to maximize genetic diversity between individuals. However, all outbred stocks start with a limited population and therefore are not nearly as heterogeneous as populations in the wild. Animals from an outbred stock will share common characteristics.
The genetic background of animals used in research is of critical importance, as it can have profound effects on the observed phenotype. A phenotype can be profound in one strain and not affect another. In addition, there are cases in which the phenotype of interest was found to be an artifact of a contributing background strain.
Genetic background analysis is commonly done one of two ways. Microsatellite analysis begins with the identification of regions of non-coding DNA containing short repeats. Each repeat motif is commonly two, three, or four base pairs in size, and the number of repeats is highly variable between different strains. Figure 1 depicts several markers from Elchrom Scientific’s (www.elchcom.com) GenoMouse™ microsatellite panel. Single nucleotide polymorphisms (SNPs) are regions of the mouse genome in which two strains differ by a single base pair. Both methods of genome scanning have advantages and disadvantages depending on the application.
Microsatellite polymorphisms can arise through replication slippage, unequal crossing over, or mutations extending or interrupting a series of repeats, whereas SNPs arise via point mutations.
As a result, new microsatellite variations arise more frequently than new SNP variations. However, the absolute number of SNP differences between strains is about a thousand-fold higher than microsatellite differences. Thus, SNP and microsatellite analyses can provide complementary information, and each is better suited for some tasks than others.
In humans, the HapMap project is attempting to delineate SNP differences between different populations with the goal of determining which SNPs are related to disease states. Because of the enormous number of SNPs in the genome, they are grouped into haplotypes. A haplotype is defined as a block of DNA that is inherited as a single unit, in which case one or a few SNPs can serve as markers for a large stretch of DNA. How applicable haplotype groups are throughout the human population is a matter of some debate.
For forensic identification and paternity testing, microsatellites are used. This enables many fewer loci to be examined, as each locus can exhibit a variety of repeat lengths. The FBI CODIS system requires just 13 microsatellite markers for a standard analysis.
Options for genome scanning include in-house screening, or core or commercial services such as Elchrom Scientific’s GenoMouse.
Misbreedings and animal mix-ups will inevitably occur, even in the best-run animal facilities. Figure 2 illustrates such a situation. Undetected, such mistakes can sabotage experiments, wasting huge amounts of money and time. Good husbandry demands routine genetic monitoring of breeding stocks. Microsatellite analysis is advantageous here, because if a contamination is detected, the sizes of the aberrant microsatellite loci can often provide information on the contaminating strain.
Because the background strain can play a major role in phenotype, it is important to examine test animals on a uniform genetic background, and ideally on several genetic backgrounds. Backcrossing, the breeding process used to transfer a gene or allele of interest from one strain to another, is a slow and laborious process, taking approximately three years. By analyzing the genomes of backcrossed offspring and selecting the animals with the highest percentage of the desired background strain, it is possible to cut the time required in half.
Mathematical modeling shows that markers spaced approximately 15 cM apart are effective, and finer mapping does not provide significantly greater speed. GenoMouse uses 96 markers, six on chromosome 1, the largest chromosome, and five markers on all the other autosomes. The sex chromosomes can usually be fixed by breeding rather than selection. The huge datasets generated by SNP analysis are unwieldy for this application.
Gene Maping and Quantitative Locus Analysis
Quantitative traits are those traits that show a continuous variation over a range of phenotypes. The phenotype is composed of the combined effects of multiple genes, some with large effects, some with small. To identify the contributing genes, animals from a genetically heterogeneous population are selected for extremes of the trait of interest. They would then be analyzed by genome scanning in order to identify a locus that segregates with the trait.
Frequently a known candidate gene is present in this region, but sometimes not. The spacing of mouse microsatellite markers, at about one every 100 kb, can be sufficient to identify the gene. SNP markers occur approximately once every kb. It is straightforward to begin a gene hunt with microsatellite markers and then switch to SNPs once the region has been sufficiently defined.
One significant advantage of microsatellite analysis is its robustness. The difficult part of the assay is identifying primer sets that amplify well, are well-spaced across all the chromosomes, and show polymorphisms with a size difference sufficient for accurate resolution with the chosen analysis technique. Primer panels may be purchased commercially or developed in-house (a significant research and development undertaking).
Once the primer sets are in hand, microsatellite analysis can be reassuringly straightforward. Markers are amplified via standard PCR methods and usually sized on gels. Agarose gels are common, although sometimes polyacrylamide gels are chosen for their finer resolution. Elchrom Scientific has a series of hydrogels, which offer clear bands and discrimination of bands with as small a distance as a single base pair. These provide agarose gel electrophoresis with precision equal to to that of polyacrylamide gels.
Data interpretation of microsatellite assays is straightforward and based on visual identification of the polymorphisms. Unlike microarray data, statistical analysis is not required, and the assay is highly reproducible between different researchers and laboratories.
The mouse genome differs significantly from the human genome. Some applications that are SNP-based for the human genome may be better approached in the mouse by microsatellite analysis, as mice have two to three times as many microsatellite loci as humans. Mice also have much longer SNP deserts. Thus mouse researchers should evaluate their needs carefully and not rely on human precedent in choosing whether to genotype with SNPs or microsatellites.